1. Field of the invention
The present invention relates to digital image data processing apparatus and method and, more particularly, to video recording and reproducing apparatus and method for transmitting digital image data by means of progressive component signals.
2. Description of the prior art
The D-1 VCR standard for recording a digitally coded video signal by component signals, and the D-2 VCR and D-3 VCR standards for recording composite signals, are currently used for recording and reproducing digital video signals. These standards sample the video signal at a sampling rate sufficiently higher than the video signal band, and quantize the sampled signal at an 8-bit rate for recording to magnetic tape.
Each of these methods, however, records a 4:2:2 signal obtained by sampling the current broadcasting-standard NTSC or PAL signal, or luminance signal (Y) at 13.5 MHz, and the color difference signals (Pb, Pr) at 6.75 MHz.
Two new next-generation broadcasting standards are also being developed, progressive component signal and HD. To handle the data quantity, which is several times that of conventional broadcasting methods, transmitted by these new digital formats, systems are being developed for compressing the video data with high efficiency coding using the correlations between data within the video signal to enable recording by current VCRs. While progressive component signal cameras and monitors have already been developed for use in the progressive component signal method, a compatible VCR has yet to be developed. As a result, two D-1 VCR units, for example, are currently linked and used as a progressive component signal VCR.
A Y, Pb, Pr progressive component video signal with 525 scanning lines per field (hereinafter "525p video signal") is an 8:4:4 digitized signal with a sampling frequency twice the sampling frequency of the 4:2:2 member of the family as defined by Recommendation ITU-R.601. The luminance signal sampling frequency in this 8:4:4 signal is therefore 27 MHz, the color difference signal sampling frequency is 13.5 MHz, and the samples are quantized at 10 bits per sample.
Furthermore, according to the Draft SMPTE Standard for Television (4:2:2:4 and 2=(4:2:2) Bit-serial Digital Interface for progressive 483 active line scanning systems; Aug. 23/1, 1994), this main 8:4:4 signal is divided for transmission into two interlaced signals, a main signal consisting of a luminance signal (Y') and two color difference signals (Pd', Pr'), and a sub signal consisting of a luminance signal (Y") and two color difference signals (Pd", Pr"), interleaved on a line by line basis. The main and sub signals are each 4:2:2 525-line interlaced signals ("525i" below) having 525 lines per frame. The signal combining the main and sub signals is referred to as a 4:2:2:4:2:2 signal.
If the 4:2:2:4:2:2 signal is Y':Pb':Pr':Y":Pb":Pr", the pairs Y' and Y", Pb' and Pb", and Pr' and Pr" express the 525p luminance and color difference signals. When the color difference component Pb" and Pr" sub signals are omitted, the resulting 4:2:2:4 signal is a reduced chrominance bandwidth signal.
The vertical frequency bandwidth of the color difference signals only is limited by a vertical line sampling filter. The 8:4:4 signal in which the vertical frequency band was limited by the vertical line sampling filter is the "8:4v:4v signal." The 4:2:2:4:2:2 signal and 4:2:2:4 signal are generated by dividing the 8:4v:4v signal on a line by line basis. Three types of 1:2:1 line sampling filters, the structure of which is shown in FIG. 5, are currently being studied.
Systems for compressing the 8:4:4 signal for recording to a current 4:2:2 VCR are being considered. Systems for compressing to 2/3 and recording the 4:2:2:4 signal in which the vertical frequency bandwidth of the color difference signals is limited are also being considered.
The digital signal recording apparatus described in Japanese patent Laid-open Publication kokai H2-14468 (1990-14468) converts a double-scan video signal to two interlaced video signals and compresses the interlaced signals 1/2 for recording to a current 4:2:2 VCR. What it does not describe are specifics about the compression method, or the slow-speed reproduction capabilities of the proposed system.
Furthermore, said proposed digital signal recording apparatus applies signal compression after converting the double-scan, 525 lines/field video signal to two interlaced video signals, resulting in 262 or 263 scanning lines per field in each of the interlaced signals. When one of the two interlaced video signals is the main signal and the other is the sub signal, e.g., if there are 262 scanning lines in the main signal of the odd fields and 263 scanning lines in the sub signal, there will be 263 scanning lines in the main signal of the even fields and 262 scanning lines in the sub signal of the even fields. If the valid 244 lines, for example, of these main and sub signals are separately compressed, the first line of the valid data will be offset 0.5 line between the main and sub signals, and this 0.5 line offset will differ in the even and odd fields. FIGS. 8A and 8B show the structure of the main and sub signals, respectively.
As shown in FIGS. 8A and 8B, when a video signal having 525 scanning lines per field is divided between main and sub signals, the main signal (real lines in FIG. 8A) consists of the even numbered lines (2, 4, . . . 524), and the sub signal (real lines in FIG. 8B) consists of the odd numbered lines (1, 3, . . . 525), in the odd fields; but, in the even fields, the main signal (dotted lines in FIG. 8A) consists of the odd numbered lines (1, 3, . . . 525), and the sub signal (dotted lines in FIG. 8B) consists of the even numbered lines (2, 4, . . . 524). As a result, when the interlaced main and sub signals thus divided are compressed and recorded by separately compressing the 244 main signal scanning lines and the 244 sub signal scanning lines from line 20 to line 507, for example, the main signal of field 1 consists of lines 20, 22, . . . 506, and the sub signal of field 1 consists of lines 21, 23, . . . 507. In field 2, the main signal consists of lines 21, 23, . . . 507, and the sub signal consists of lines 20, 22, . . . 506. When reproducing a tape thus recorded, the source signal can be reproduced during normal reproduction modes wherein the output field and the field on the tape coincide. During slow-speed reproduction modes in which the output field and the field on tape do not necessarily coincide, however, the output field may be an odd field while the field on tape is an even field, in which case the line numbers of the output main signal are 21, 23, . . . 507 while the line numbers of the output sub signal are 20, 22, . . . 506. As a result, the odd field progressive component signal combining these main and sub signals will be output in the line sequence 21, 20, 23, 22, . . . 507, 506, and the normal progressive component signal cannot be generated.